Ethynylation of 17-keto steroids to produce commercially important 17.alpha.-ethynyl-17.beta.-hydroxy steroids is well known to those skilled in the art. See, for example, U.S. Pat. Nos. 2,272,131, 2,843,609, 2,723,280, 2,877,240, 3,470,217, 4,041,055, Steroids by Fieser and Fieser, Reinhold Publishing Co., New York, 1959, pp. 557-591, and J. Am. Chem. Soc. 78, 2477 (1956).
The general method of ethynylation is to react the 17-keto steroid with dipotassium acetylide. The advantage of the dipotassium acetylide process is that it can be used with .DELTA..sup.4 -3-keto steroids without having to protect the 3-keto group. However, that procedure cannot be used with 16.alpha.-methyl-17-keto, 16.beta.-methyl-17-keto or 16-methylene-17-keto steroids for well known reasons. Commercially the ethynylation of 16.alpha.- or 16.beta.-methyl- as well as 16-methylene-17-keto steroids is important because the 17.alpha.-ethynyl-17.beta.-hydroxy-16.alpha.-methyl, 17.alpha.-ethynyl-17.beta.-hydroxy-16.beta.-methyl and 17.alpha.-ethynyl-17.beta.-hydroxy-16-methylene steroids can be transformed to dexamethasone, betamethasone and melengestrol acetate.
Metallo-acetylides other than dipotassium acetylide are known. Monosodium acetylide is known, see U.S. Pat. No. 3,470,217 and R. J. Tedeschi, et al., J. Org. Chem. 34, 435 (1969). Mono- and bis-magnesium acetylides are known, see L. Skattebol, et al., J. Chem. Soc. 4765 (1956). Although the use of magnesio-acetylides has been reported for 17.alpha.-ethynyl introduction, substantial dimer formation results with both mono- and bis-magnesioacetylides, see U.S. Pat. No. 3,704,253.
Lithioacetylide reagents exhibit substantially different reactivity in many cases from other metallo-acetylides. This fact and the ready availability of n-butyllithium has resulted in the extensive use of these reagents in syntheses. The covalent nature of the carbon-lithium bond has been the subject of many theoretical and experimental investigations, see, for example, J. B. Moffat, J. Mol. Structure 42, 251 (1977) and A. Streitwieser, et al., J. Am. Chem. Soc. 98, 4778 (1976).
M. M. Midland in J. Org. Chem. 40, 2250 (1975) reported reacting n-butyllithium with acetylene in THF at low temperature (&lt;-70.degree.) and in dilute solution to produce monolithium acetylide. Monolithium acetylide is a valuable reagent for the preparation of ethynyl carbinols and terminal acetylenes, see Fieser and Fieser, Reagents for Organic Synthesis, Vol. 1, Wiley, New York, 1967, p 573. Midland found that warming or attempting to generate a more concentrated solution resulted in disproportionation to the insoluble dilithium acetylide and acetylene. This disproportionation is an important disadvantage and occurs in the absence of a complexing agent, see Corbellini et al., Chem. Ind. (Milan) 42, 251 (1960) and Chem. Abstr. 54, 19250 (1960). To reduce or prevent the disproportionation the monolithium acetylide is usually prepared in liquid ammonia, which presumably serves as an appropriate complexing agent. An amine such as ethylenediamine can also be used to stabilize monolithium acetylide. Ethylenediamine so greatly stabilizes monolithium acetylide that monolithium acetylide is sold commercially as an ethylenediamine complex. Ethylenediamine while stabilizing monolithium acetylide to the point it can be sold commercially actually reduces the reactivity to the point it is not useful for many ethynylation procedures.
The addition of a lithiated acetylene species to a 16.beta.-methyl-17-ketone in 92% yield was reported by G. Neef, et al., in J. Org. Chem. 43, 4679 (1978) without giving any experimental data but stating, "The ethynylations were performed according to the procedure of Phillips . . . " The procedure of Phillips is set forth in J. Med. Chem. 11, 924 (1968). The results reported by Neef could not be reproduced; the Phillips procedure consistently gave large amounts (greater than 20%) of irreversible enolization. Neef also reported the addition of a lithiated acetylene species to a 16.alpha.-methyl-17-keto steroid. The yield Neef reported (72%) is more in keeping with the observed irreversible enolization.
U.S. Pat. No. 4,055,562 used monolithium acetylide to ethynylate 17-keto steroids unsubstituted in the C.sub.16 position. The monolithium acetylide was prepared by bubbling acetylene into THF held at -70.degree. under anhydrous conditions followed by addition of butyllithium. The 17-keto steroid was added to the unstabilized monolithium acetylide and the mixture stirred for 3 hr at -70.degree. to produce the 17.alpha.-ethynyl-17.beta.-hydroxy steroid product.
U.S. Pat. No. 4,320,236 discloses the use of a monolithium acetylide-ammonia complex (which is well known to those skilled in the art) to ethynylate ketones at below about 30.degree.. The examples in U.S. Pat. No. 4,320,236 disclose ethynylation reaction temperatures of -50.degree. to 10.degree.. The unsaturated acyclic ketones ethynylated in U.S. Pat. No. 4,320,236 are very reactive whereas the steroidal 16-methyl-17-ketones are highly substituted sterically hindered cyclopentanones and therefore much less reactive.
The process of the present invention uses monolithium acetylide and obtains 17.alpha.-ethynylation in yields greater than 85% with no detectable isomerization of the 16.alpha.- or 16.beta.-methyl groups, indicating less than 2% epimerization.